Codex pricing: ChatGPT plans, usage limits and what uses them
Which ChatGPT plans include Codex, how many messages each model allows, how API-key use is billed, and a measured test of which model passed a coding task for the least.
Codex comes with your ChatGPT plan, or pay per token with an API key
Free / Go | Plus | Pro | Business | API key | |
|---|---|---|---|---|---|
| Codex included | Limited trial | ✓ | ✓ | ✓ | ✓ |
| Codex cloud, GitHub code review, Slack | -- | ✓ | ✓ | See OpenAI's plan table | -- |
| Five-hour usage limit | Not published | Yes | None currently | Yes | No plan limit |
| How extra usage is paid | -- | Prepaid ChatGPT credits | Prepaid ChatGPT credits | Prepaid workspace credits | Per-token API rates |
How many messages each model allows, per five hours
Local messages per five hours on Plus and Standard Business (OpenAI's estimates, not fixed limits; checked 9 Nov 2026)
Current OpenAI model rates, per million tokens
| Model | Input | Output | Batch in | Batch out | Cache read |
|---|---|---|---|---|---|
| GPT-4.1 | $2.00 | $8.00 | $1.00 | $4.00 | $0.50 |
| GPT-4.1 Mini | $0.40 | $1.60 | $0.20 | $0.80 | $0.10 |
| GPT-4.1 Nano | $0.10 | $0.40 | $0.05 | $0.20 | $0.03 |
| GPT-4o | $2.50 | $10.00 | $1.25 | $5.00 | $1.25 |
| GPT-4o Mini | $0.15 | $0.60 | $0.07 | $0.30 | $0.07 |
| GPT-5 | $1.25 | $10.00 | $0.62 | $5.00 | $0.12 |
| GPT-5 Mini | $0.25 | $2.00 | $0.12 | $1.00 | $0.03 |
| GPT-5.1 | $1.25 | $10.00 | $0.62 | $5.00 | $0.12 |
| GPT-6 Astra | $10.00 | $50.00 | $5.00 | $25.00 | $1.00 |
| GPT-6.1 Sol | $2.00 | $10.00 | $1.00 | $5.00 | $0.10 |
| GPT-6.1 Terra | $1.00 | $5.00 | $0.50 | $2.50 | $0.05 |
| o3 | $2.00 | $8.00 | $1.00 | $4.00 | $0.50 |
| o3 Mini | $1.10 | $4.40 | $0.55 | $2.20 | $0.55 |
| o4 Mini | $1.10 | $4.40 | $0.55 | $2.20 | $0.28 |
All prices in USD per 1M tokens. Batch runs asynchronously at a reduced rate; cache read applies to reused input tokens. A dash means the provider does not publish that rate.
With an API key, Codex is billed at these live per-token rates. Full table: OpenAI API pricing.
Best model for Codex: on our task, GPT-6.1 Terra passed for half of Sol's cost
Cost per 1,000 answers, same coding task (evaluation 75cdd5e8, 9 Nov 2026; all scored 100)
Have your agent test which OpenAI model to use
Fast mode uses your included usage 2.5 times faster
How fast each speed mode uses your allowance, vs Standard (OpenAI, checked 9 Nov 2026)
Find the cheapest OpenAI model that passes your task
Score Terra, Sol and Astra on your own task, from your agent or the dashboard. On Pro and above.
Turns and tool calls eat a coding agent's allowance
Tool calls vs credits, one point per step of a Kiro workflow of ours (agent analysis, 4 Nov 2026)
On Pro: for Codex users
- Score OpenAI models on your own task
- Terra, Sol and Astra compared on cost, score and speed
- Step-by-step analysis of Kiro workflows and Claude Code sessions
- Scheduled re-tests flag when an OpenAI model's quality slips
Use it next to Codex, over MCP (the protocol agents use to call outside tools)
- Your agent runs the tests and reports back
- Connect it to Codex as an MCP server
- Agent analysis for Kiro workflows and Claude Code sessions, the trace formats it reads today
$129/mo $10 for your first week
Then $129/mo. Cancel anytime.
Plan contents, usage estimates and speed multipliers are from OpenAI's Codex pages, checked 9 November 2026; OpenAI lists plan prices on its own pricing page. Per-token rates come from our live model registry. The measured result is one evaluation of ours; your own task will score differently.
Questions
Is Codex free?
What are the Codex usage limits?
Plan or API key: which should I use?
How do I make my Codex usage last longer?
Score models on your own coding task and analyse your agent workflows from your agent over MCP, on Pro and above: see how agent evaluation works.